Research map / 3D vision / Gaussian splatting
Liberated-GS: 3D Gaussian Splatting Independent from SfM Point Clouds
Weihong Pan, Xiaoyu Zhang, Hongjia Zhai, Xiaojun Xiang, Hanqing Jiang, Guofeng Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add depth alignment (ablation) | Liberated-GS base model · Mip-NeRF360 | PSNR: 27.079 -> 27.285 | Improved |
| add local Gaussian for unseen areas after depth alignment (ablation) | Liberated-GS · Mip-NeRF360 | PSNR: 27.285 -> 27.588 | Improved |
| add local Gaussian (ablation) | Liberated-GS · Mip-NeRF360 | LPIPS: 0.190 -> 0.187 | Improved |
| use ensembled depth instead of aligned depth (ablation) | Liberated-GS · Mip-NeRF360 | PSNR: 27.524 -> 27.588 | Improved |
| use per-object affine aligned depth instead of scale-consistency estimated depth (ablation) | Liberated-GS · Mip-NeRF360 | PSNR: 27.390 -> 27.524 | Improved |
| use rendered depth only instead of ensembled depth (ablation) | Liberated-GS · Mip-NeRF360 | PSNR: 27.588 -> 26.596 | Hurt |
| replace random init with Liberated-GS init | Mini-Splatting · OMMO | PSNR: 25.55 -> 27.29 | Improved |
| replace SfM init with Liberated-GS init | Mini-Splatting · OMMO | PSNR: 26.87 -> 27.29 | Improved |
| replace SfM init with Liberated-GS init | 3DGS-MCMC · OMMO | PSNR: 28.34 -> 28.50 | Improved |
| replace random init with Liberated-GS init | 3DGS-MCMC · OMMO | PSNR: 27.75 -> 28.50 | Improved |
| replace SfM init with Liberated-GS init | 3DGS · OMMO | PSNR: 27.79 -> 28.01 | Improved |
| replace SfM init with Liberated-GS init | 3DGS · Tanks and Temples | LPIPS: 0.183 -> 0.140 | Improved |
Closest papers by meaning
- SfM-Free 3D Gaussian Splatting via Hierarchical Training (CVPR 2025 · similarity 0.93)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.92)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.92)
- SplatFormer: Point Transformer for Robust 3D Gaussian Splatting (ICLR 2025 · similarity 0.92)
- EAP-GS: Efficient Augmentation of Pointcloud for 3D Gaussian Splatting in Few-shot Scene Reconstruction (CVPR 2025 · similarity 0.92)
- PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
